JPS625485A - Optical character recognizing device - Google Patents

Abstract

PURPOSE:To simplify an instruction program for instructing the read of a slip by providing a means which recognizes color instruction information and a means which performs a specific recognition processing on the basis of the color instruction information recognized by said recognizing means. CONSTITUTION:A color scanner part 12 reads the color images of the slip and outputs the color image information to a color image fetch part 13. The fetch part 13 retrieves a color frame from the color image information and fetches an image as to the frame of a necessary color, thereby sending image data to a character image output part 7. Simultaneously, the position of the color frame is calculated and sent to a color character recognizing part 14 together with the color frame information. The recognizing part 14 recognizes the character in the color frame according to the program 15 on the basis of the color frame position information and color frame information from the fetch part 13 and outputs the character to an output part 7. The read slip is discharged to a stacker 8 according to an output control program 16 and image information and character information are outputted to an output medium 9.

Description

[Detailed description of the invention] [Technical field of invention] The present invention relates to an optical character recognition device.

[Technical background of the invention]

Conventional forms recognized by optical character recognition devices are
output, input item names, and text entry frames in dropout color. A dropout color is a color that has low detection sensitivity by an optical character recognition device and is read when nothing is written on it. Write letters and numbers on this form with a pencil or ballpoint pen. When this filled-in form is read by an optical character recognition device, only black characters, numbers, etc. are recognized.

A conventional optical character recognition device is shown in FIG. Hopper part 1
The form placed thereon is transported, and the scanner section 2 reads the black pattern on the form. Image information of the read black pattern is captured by the image capture unit 3,
Character information is input to the character recognition section 4. image information,
Character information is distinguished by instructions from the form standard instruction program 5. The form standard instruction program 5 also performs other instructions regarding character types (numbers, alphanumeric kana, kanji, etc.). Images and characters from the image capture unit 3 and character recognition unit 4 are outputted from the character/image output unit 7 under the control of the output control program 60, and are outputted from the magnetic tape device 9a, floppy disk device 9b1, external line 9C, etc. It is sent to medium 9. The read forms are stored in the stacker section 8.

A specific example of a conventional form is shown in FIG. 7(a). This form is for creating a customer list, ■ Form identification number (ID)
It consists of a column, ■Customer name column, ■Furigana column, and ■Message number column. In order to read this form, contents as shown in FIG. 7(b) are programmed into the form standard instruction program 5. In other words, ■The form identification number (ID) a starts from the position of X mm, V1 mm of the form, has 3 characters, and is a number; ■The customer name field is the x mm, y, , starting from the mm position, the number of characters is 9, and it must be in kanji.
It is determined that it starts from the position of mm, y3mm, has 13 characters, is in katakana, etc.

The character m1 section 4 recognizes characters based on instruction information such as the position of each column, the number of characters, and the type of characters provided by the form standard instruction program 5.

Another specific example of the form is shown in FIG. 8(a). This form is ■
It consists of a slip 2 column, ■a item column, ■b item column, and ■C item column. In order to read this form, contents as shown in FIG. 8(b) are programmed into the form standard instruction program 5. In other words, the image to be captured is
mrn.

Starting from the position of y mm, an area of length xj 1 mm (heading I [ll) and ■ X 2 mm of the form,
It is determined that the area starts from the position of '12 mm and has a length of xf12 mm (all C items). The rare character image is output to the rare character image output section 7.

Still another specific example of the form is shown in FIG. 9(a).

In this specific example, a plurality of types of forms are input and classified by form type. The type of form is written in a two-digit number at a predetermined position on each form. In the form standard instruction program 5 and output control program 6, as shown in FIG. 9(b), the field in which the form type is written is xmm of the form.

It is determined that starting from the position ymm, the number of characters is 2, that it is a number, and that the form is classified into #10 to #12 sorter pockets, normal pockets, and reject pockets according to the number of the form type.

The character recognition unit 4 recognizes the number indicating the type of form using the form standard instruction program 5, and classifies the form based on the recognition result.

[Problems with background technology]

As mentioned above, parts that are not recognized in conventional forms need to be printed in dropout color. However, there is a problem in that special technology is required to print dropout colors that cannot be read by optical character recognition devices. In addition, the dropout color is not unified for all openings of optical character recognition devices, and there is a problem in that a color that is set as a dropout color in another model may be read. In order to prevent this, it is possible to lower the PC8 value of the dropout color, but there is a problem in that lowering the PC3 value makes it difficult for the human eye to see.

Furthermore, when reading a form, it is necessary to program the reading position, character type, etc. for each form in advance, as shown in the above-mentioned specific examples. In this program, as the number of items in the form and the types of forms increase, the minute S increases in proportion to the number of items and types of forms, so
The problem was that the work became difficult. Furthermore, if the layout of a form is changed, the program must be changed accordingly.

Furthermore, conventionally, it is necessary to write a form identification number on a form in order to distinguish the type of form. This form identification number was not original data and was wasted.

Furthermore, in conventional optical character recognition devices, it is necessary to leave a certain distance between characters and between lines, which causes the problem that the document size becomes larger than necessary.

[Purpose of the invention]

The present invention has been made in consideration of the above circumstances, and it is an object of the present invention to provide an optical character recognition device that simplifies an instruction program for instructing reading of a form and eliminates the need to change the instruction program even if the layout of the form is changed. With the goal.

[Summary of the invention]

To achieve the above object, an optical character recognition device according to the present invention is characterized in that a color recognition section recognizes color instruction information on a form, and recognizes the document based on this color instruction information.

[Embodiments of the invention]

An optical character recognition device according to an embodiment of the present invention is shown in FIG. A form to be read is set in the hopper section 1.

The form set in the hopper section 1 is transported by a transport mechanism (not shown).
and sent to the color scanner section 12. The color scanner unit 12 reads color images (including black) on the form. The color image capture unit 13 and color character recognition unit 14 input color image information and color character information from the color scanner unit 12.

The color-specific reading instruction program 15 determines instruction information such as the reading position of the form and the type of characters, the details of which will be described later. The character image output unit 7 outputs image data and character data from the color image capture unit 13 and color character recognition unit 14 under the control of the output control program 16.

The output image data and character data are sent to an output medium 9 such as a magnetic tape Vt storage 9a, a floppy disk device 9b, and an external line 90.

The stacker section 8 stores the read forms.

Next, the operation will be explained using FIG. 2. First, before a form reading operation, a program for a form to be read is stored in the reading instruction program 15 and the output control program 16 (step 1). The form shown in FIG. 3(a) will be explained as a specific example.

This form is used to create a customer list, and is unique in that the frames in each column are color-coded. In other words, ■The form name ("Omiya Ledger") column in the black frame, ■The customer name column in the red frame, ■The furigana column in the front frame, ■The customer number column in Yellow, ■The family composition column in Yellow, ■Yellow It consists of a postal code field in a frame and an address field in a red frame. For this V&vote, reading instruction 70 grams 1
5, program the contents shown in FIG. 3(b). That is, it is determined that: ■ Kanji are in the red frame, ■ Katakana is in the blue frame, ■ Numbers are in the yellow frame, and ■ Other color frames do not need to be read.

Next, the color image of the form is read in by the color scanner unit 12, and the color image information is transferred to the color image capture unit 1.
3 (step 2). Color image capture section 1
In step 3, a color frame is searched from the color image information and the color of the frame is recognized. Then, according to the reading instruction program 5,
An image is captured for the frame of the necessary color, and the image data is sent to the character image output section 7. At the same time, the position of the color frame is calculated and sent to the color character recognition unit 14 together with the color frame information (step 3).

The color character recognition unit 14 recognizes the characters within the color frame according to the reading instruction program 15 based on the color frame position information and color frame information from the color image capture unit 13 (step 4).
). In this specific example, the text inside the red frame is kanji, so the customer name and address fields are recognized in kanji mode, and the text in the front frame is katakana, so the furigana field is recognized in katakana mode.
Since the numbers within the yellow frame are numbers, the individual number field, family structure field, and postal code field are recognized in numerical mode. Note that since there is no need to read the area within the black frame, the color image capture unit 13 may not read it, or the color character recognition unit 14 may not read it.
You may choose not to recognize it. The recognized character information is output to the character image output section 7.

The read form is discharged to the stacker unit 8 according to the output control program 6 (step 5). The image information and character information are then output to the output medium 9 according to the output control program 16 (step 6).

Another specific example of the form is shown in FIG. 4(a). This form is ■
It consists of a slip name column with a black frame, ■ an item a column with a red frame, ■ an item b column with a front frame, and ■ an item C column with a yellow frame. In order to read this form, the reading instruction program 15 determines to capture the image inside the black frame and the yellow frame, as shown in FIG. 4(b).

As a result, when capturing an image in step 3, the color image capturing unit 13 captures only the black frame in the slip name column and the yellow frame in the C item column and outputs the image to the character image output unit 7.

Another specific example of the form is shown in FIG. 5(a). In this specific example, a plurality of types of forms are manually classified by form type. As shown in FIG. 5(a), the color of the frame in the form name field is changed depending on the type of form. That is, a slip has a frame of color a, a slip of b has a frame of color b, a slip of C has a frame of color 0, and a slip of d has a frame of color d. As shown in FIG. 5(b), the reading instruction program 15 and the output control program 16 output the frame to the sorter pocket #10 if the color a is the color of the frame, and the sorter pocket #11 if the frame color is the color b. If the color is 0, it is to be discharged to the #12 sorter pocket, if the color is d, it is to be discharged to the normal pocket, and if it is any other color, it is to be discharged to the reject pocket. As a result, when a form is output, the form is classified according to the color of the frame in the form name field.

In this way, according to this embodiment, the instruction information for the form is given by the color of the frame, so the layout of the form can be changed freely as long as the color of the frame is not changed. Furthermore, since there is no need to specify the position of each column, the program becomes simpler. Furthermore, if the frame colors are unified, there is no need to create a program for each form.

Furthermore, if the forms are classified by the color of the frame, there is no need to write the form identification number on the form.

By simply enclosing each column with a color frame, you can capture an image of that column, so you can freely set the line spacing and character spacing of the form. Therefore, it is possible to easily design the form and to reduce the size of the form.

The present invention is not limited to the above embodiments, and various modifications are possible.

For example, the color of the paper may be changed depending on the type of form, and the color of the paper may be recognized and classified. The present invention can also be applied to a color scanner of an image processing device.

〔Effect of the invention〕

As described above, according to the present invention, a simple reading instruction program is sufficient, and there is no need to change the instruction program even if the layout of a form is changed. In addition, forms can be designed with free pitch, making form design easy and downsizing possible.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an optical character recognition device according to an embodiment of the present invention, FIG. 2 is a flowchart showing the operation of the optical character recognition device, and FIGS. 3 to 5 are a block diagram of an optical character recognition device according to an embodiment of the present invention. FIG. 6 is a block diagram of a conventional optical character recognition device, and FIGS. 7 to 9 show specific examples of a form and reading instruction program in the same optical character recognition device. It is a figure which shows an example. 1...Hopper section, 12...Color scanner section, 13
...Color image capture unit, 14...Color character recognition unit, 15...Reading instruction program, 16...Output control program, 7...Character image output unit, 8...
- Stacker section, 9...output medium. Applicant's agent Inomata Seiki 3 (α) Chi 4 district (b3 Oni 5 diagram Oni 6 diagram (α) Keiryo map (α) Guan 6 diagram Pond 9 diagram

Claims (1)

[Claims] An optical character recognition device that recognizes letters, numbers, etc. written on a form includes a color recognition unit that recognizes color instruction information that displays instructions on the form in color, and color instruction information recognized by the color recognition unit. What is claimed is: 1. An optical character recognition device comprising: a recognition processing unit that performs predetermined recognition processing based on the above.